Fig 1.
Stimulation of P-TEFb assembly in memory CD4+ T cells by TCR co-stimulation and PKC agonists.
(A) Proposed scheme for the biogenesis of P-TEFb in memory CD4+ T cells and its exchange from 7SK snRNP to assemble the super elongation complex (SEC) on proviral HIV. Posttranscriptional synthesis of cyclin T1 (CycT1) initiated immediately upon TCR co-stimulation serves as a trigger for the heterodimeric assembly of P-TEFb and T-loop phosphorylation of CDK9 kinase at Thr186. Assembled P-TEFb is translocated into the nucleus where it is associated with HEXIM1 and 7SK snRNA to form 7SK snRNP. Exchange of P-TEFb from 7SK snRNP is facilitated by the displacement of HEXIM1 by Tat and also T-loop phosphorylation of CDK9 at Ser175 leading to formation of the super elongation complex (SEC) containing transcriptionally active P-TEFb. This complex is then loaded onto TAR RNA to stimulate proviral transcription elongation. Signaling pathways responsible for the biogenesis of P-TEFb leading up to formation of transcriptionally active P-TEFb will be examined in this study. (B) Assessment of active P-TEFb expression by immunofluorescence flow cytometry as measured by monitoring the co-expression of CycT1 and pSer175 CDK9. (C) Flow cytometry analysis of the intracellular fluorescence immunostaining of total CDK9, cyclin T1 (CycT1), pThr186 CDK9 and pSer175 CDK9 from at least five different experiments performed using either healthy donor-derived memory T cells (black) or in vitro polarized primary Th17 cells (red) that were activated or not for 18 h through the TCR with anti-CD3/anti-CD28 Dynabeads. Statistical significance (p values) was calculated using a two-tailed Student’s t test. (D) Assessment of stimuli capable of inducing P-TEFb expression in primary T cells. The graph shows immunofluorescence flow cytometry measurements of the coordinate induction of CycT1 and pSer175 CDK9 expression in memory T cells (black) derived from at least four different healthy donors or in vitro polarized primary Th17 cells (red). Cells were subjected to 24 h treatment with the indicated stimuli prior to immunofluorescence staining for CycT1 and pSer175 CDK9. The stimuli were used at the following concentrations: a) 1:1 bead-to-cell ratio of anti-CD3/anti-CD28 Dynabeads; b) 1 μg/ml anti-CD3 antibody; c) 1 μg/ml ionomycin; d) 50 nM ingenol; e) 1 μM prostratin; f) 50 ng/ml PMA; g) 10 ng/ml TNF-α; h) 500 nM SAHA.
Fig 2.
Imaging the nuclear assembly of P-TEFb and 7SK snRNP in activated memory CD4+ T cells.
(A) Subcellular expression of P-TEFb subunits in memory CD4+ T cells before and after T-cell receptor (TCR) activation. Cells were stimulated or not with a soluble cocktail of anti-CD3/anti-CD28 antibodies (500 ng/ml anti-CD3 and 1 μg/ml anti-CD28) for 4 or 24 h prior to dual immunostaining for CDK9 and CycT1. Images were captured at 100X using a deconvolution microscope. Scale bar: 10 μm. (B) Visualization of the expression and subcellular localization of 7SK snRNA and HEXIM1 in memory CD4+ T cells. These imaging experiments were carried out by combining RNA FISH with immunofluorescence staining of memory T cells that were either unstimulated or were activated through the TCR for 2 or 24 h with anti-CD3/anti-CD28 Dynabeads at a 1:1 bead-to-cell ratio, or were challenged with a combination of 1 μM prostratin and 1 μg/ml ionomycin. Images were captured at 100X by deconvolution microscopy. Scale bar: 5 μm.
Fig 3.
Transcriptome analysis of P-TEFb, 7SK snRNP and SEC expression in primary QUECEL subsets and memory CD4+ T cells.
Quiescent CD4+ T cells that had been polarized into Th1, Th2, Treg and Th17 subsets using the QUECEL procedure shown in S4 Fig were activated or not for 24 h with anti-CD3/anti-CD28 Dynabeads. Bulk RNA-seq datasets obtained using these cells were analyzed to examine the expression of P-TEFb subunits (A), 7SK snRNP components (B), and factors belonging to the super elongation complex (C). Transcripts per million (TPM) values were used to evaluate the relative abundance of transcripts under resting and activated conditions. In A, the expression of CD25 was examined as a positive control. A publicly available bulk RNA-seq dataset of primary human memory CD4+ T cells that had been activated or not through TCR co-stimulation (SRA accession SRP026389) with anti-CD3/anti-CD28 coated beads was also analyzed for these factors (Tm Expt 1 and Tm Expt 2). Statistical significance (p values) was calculated using a two-tailed Student’s t test.
Fig 4.
P-TEFb biogenesis in primary T cells does not require intracellular mobilization of calcium.
(A) Left graph; Cross-comparison of the effectiveness of 50 nM ingenol, 1 μM prostratin or 50 ng/ml PMA on their own or combined with 1 μg/ml ionomycin at inducing active P-TEFb in primary T cells measured by co-staining for CycT1 and pSer175 CDK9. Middle graph; Cross-comparison of the effectiveness of 50 nM ingenol, 1 μM prostratin or 50 ng/ml PMA on their own or combined with 1 μg/ml ionomycin at coordinately inducing the C-terminal domain Ser2 phosphorylated form of RNA polymerase II (pSer2 RNAP II CTD) and the transcriptionally active form of NF-κB (pSer529 p65). Right graph; Measurement of the effect of 50 nM ingenol, 1 μM prostratin or 50 ng/ml PMA on their own or combined with 1 μg/ml ionomycin on cell viability as assessed by flow cytometry following staining with the eFluor450 viability dye. (B) Measurement of P-TEFb (left graph) and latent HIV (right graph) reactivation following 24 h treatment of resting and polarized Th17 cells with 50 nM ingenol, 1 μM prostratin or 1 μg/ml ionomycin. The graphs show data generated using polarized Th17 cells prepared using naïve T cells from at least two different donors. (C) Proposed schemes for the promoter recruitment of RNA polymerase II (RNAP II) to proviral HIV and its phosphorylation by active P-TEFb on its C-terminal domain (pSer2 RNAP II CTD). Arrows with red question marks indicate a P-TEFb regulatory signaling pathway(s) intended to be defined by the current study.
Fig 5.
Coordinate reactivation of P-TEFb and latent HIV in primary T cells is largely independent of PKC activity.
(A) PKC-θ and pan-PKC inhibition are inefficient at repressing ingenol- or prostratin-induced expression of P-TEFb, but effective at blocking the activating phosphorylation of the p65 subunit of NF-κB at Ser529 in response to these phorbol esters. CD4+ memory T cells from four different donors and in vitro polarized primary Th17 cells were treated for 24 h with the indicated concentrations of either ingenol or prostratin in the presence or absence of either 100 nM sotrastaurin or 100 nM Ro-31-8220 prior to subjecting cells to dual immunofluorescence staining for pSer529 p65 (top graph) or for CycT1 and pSer175 CDK9 (bottom graph). (B) PKC inhibitors modestly repress TCR-mediated expression of P-TEFb in CD4+ memory T cells but are unable to block reactivation of proviral HIV in primary Th17 cells. Left graph; Measurement of P-TEFb expression by dual CycT1 and pSer175 CDK9 immunofluorescence staining of memory T cells that were TCR-activated with anti-CD3/anti-CD28 Dynabeads for 24 h in the presence or absence of treatment with either 100 nM sotrastaurin or 100 nM Ro-31-8220. Right graph; Assessment of the effect of PKC inhibition on proviral reactivation in ex vivo HIV-infected primary Th17 cells prepared from naïve CD4+ T cells belonging to three healthy donors. Latently infected cells were treated or not for 30 min with 100 nM Ro-31-8220 in two experiments or with 100 nM Sotrastaurin in three experiments prior to TCR co-stimulation with anti-CD3/anti-CD28 Dynabeads for 24 h. Thereafter, cells were analyzed by flow cytometry following immunostaining using a fluorophore-conjugated antibody towards HIV Nef. (C) PKC inhibitors are unable to repress the reactivation of latent HIV in primary Th17 cells that is in response to PKC agonists. Latently infected Th17 cells were treated or not for 30 min with either a combination of Ro-31-8220 and Gö 6983 (Donor 14 (Expt 1) and Donor 14 (Expt 2)) at 100 nM each or 100 nM Sotrastaurin (Donor 16 (Expt 1) and Donor 14 (Expt 3)) prior to TCR co-stimulation or challenge with 50 nM ingenol or 1 μM prostratin. Thereafter, cells were immunostained using a fluorophore-conjugated antibody towards HIV Nef and analyzed by flow cytometry. Statistical significance (p values) in A, B and C was calculated using a two-tailed Student’s t test.
Fig 6.
Inducible P-TEFb expression and proviral reactivation in primary T cells are mediated by the MAPK ERK pathway.
(A) Proposed scheme for the intracellular stimulation of P-TEFb expression in primary T cells by PKC agonists or TCR-generated diacylglycerol (DAG) leading up to the reactivation of HIV from latency. (B) The MEK inhibitor U0126 substantially blocks expression of active P-TEFb in primary CD4+ T cells in response to treatment with PKC agonists but modestly inhibits P-TEFb expression in response to TCR co-stimulation. Memory CD4+ or primary Th17 cells were treated or not for 30 min with U0126 at the indicated concentrations prior to TCR co-stimulation with anti-CD3/anti-CD28 Dynabeads or challenge with either 50 nM ingenol or 1 μM prostratin for 24 h. Cells were then analyzed by flow cytometry for P-TEFb expression following immunostaining using fluorophore-conjugated antibodies towards CycT1 and pSer175 CDK9. (C) U0126 substantially retards proviral reactivation in primary Th17 cells in response to TCR co-stimulation or challenge with PKC agonists. Latently infected Th17 cells prepared using naïve CD4+ T cells from two different donors were pretreated or not with 20 μM U0126 for 30 min prior to TCR co-stimulation with anti-CD3/anti-CD28 Dynabeads or challenge with 50 nM ingenol or 1 μM prostratin for 24 h. Cells were analyzed by flow cytometry following immunostaining using a fluorophore-conjugated antibody towards HIV Nef. All the p values shown in B and C were calculated using a two-tailed Student’s t test.
Fig 7.
Activated memory CD4+ T cells predominantly express the RasGRP1 isoform which rapidly localizes to the plasma membrane upon phorbol ester stimulation.
(A) Immunofluorescence microscopy of healthy donor memory CD4+ T cells with or without a 2-h stimulation with TCR Dynabeads, 50 nM ingenol, or 1 μM prostratin. Following stimulation cells were fixed in 4% formaldehyde, permeabilized, and then immunofluorescently co-stained for RasGRP1 and a constitutively expressed membrane protein Ezrin. After counterstaining with DAPI, coverslips were mounted onto glass slides and imaged using a DeltaVision deconvolution microscope at 100X. Scale bar represents a length of 5 μm. The white arrow points to an activator bead coated with anti-CD3 and anti-CD28 antibodies that was immunoreactive towards anti-mouse secondary antibody. (B) Degree of membrane colocalization of RasGRP1 and Ezrin as assessed by measuring multiple Pearson’s correlation coefficient values obtained from defined regions of interest. Selected regions of interest defined as single cells were analyzed using the softWoRx colocalization module to measure the degree of colocalization between fluorescently stained RasGRP1 and Ezrin. Statistical significance (p values) was calculated using a two-tailed Student’s t test. (C) UMAP clustering based on scRNA-seq profiling of resting unstimulated and TCR-activated memory CD4+ T cells. (D) Assessment of the expression profiles of RasGRP isoforms in memory CD4+ T cells by scRNA Drop-seq. (E) Western blotting analysis of RasGRP1 and RasGRP2 using whole cell extracts belonging to memory CD4+ T cells that were activated or not for the times indicated with anti-CD3/anti-CD28 Dynabeads. (F) Densitometry analysis of RasGRP1 and RasGRP2 expression in memory T cells. Graphed data are from three separate experiments and show relative expression of the isoforms following normalization to the corresponding expression levels of β-Actin.
Fig 8.
Inhibition of PI3K and mTORC kinases partially suppresses TCR-induced P-TEFb expression and modestly inhibits TCR-induced proviral reactivation in primary CD4+ T cells.
(A) Proposed role of the PI3K/AKT/mTOR signaling pathway in mediating the biogenesis of P-TEFb. Inhibitors that were tested in the current study are shown in black italics. (B) PI3K and mTORC kinase inhibitors partially repress TCR-induced expression of active P-TEFb and albeit with significant variability. Memory CD4+ T cells (three experiments) and resting primary Th17 cells (four experiments) were treated or not for 30 min with inhibitors targeting PI3K (LY294002), mTORC1 (Rapamycin) or both mTORC1 and mTORC2 (Torin) at the concentrations shown prior to TCR co-stimulation for 24 h. Thereafter, cells were examined for active P-TEFb expression by flow cytometry following dual immunofluorescence staining for CycT1 and pSer175 CDK9. (C) PI3K and mTORC kinase inhibitors modestly repress proviral reactivation that is in response to TCR co-stimulation. Latently infected Th17 cells prepared using naïve CD4+ T cells isolated from three healthy donors were treated or not for 30 min with inhibitors to PI3K (LY294002), mTORC1 (Rapamycin) or both mTORC1 and mTORC2 (Torin) prior to stimulating the cells through the TCR for 24 h. Cells were thereafter analyzed by flow cytometry following immunostaining using a fluorophore-conjugated antibody towards HIV Nef. Statistical significance (p values) in B and C was determined using a two-tailed Student’s t test.
Fig 9.
Combined inhibition of MEK and mTORC kinases abrogates TCR-induced P-TEFb expression and effectively suppresses TCR-mediated proviral reactivation in primary CD4+ T cells.
(A) Combined inhibition of MEK and PI3K kinases enhances the disruption of active P-TEFb expression. Healthy donor memory CD4+ T cells (black) and polarized resting primary Th17 cells (red) were treated or not for 30 min with inhibitors targeting either MEK (U0126) or PI3K (LY294002) or both at the concentrations shown prior to TCR co-stimulation for 24 h. Thereafter, cells were examined for active P-TEFb expression by flow cytometry following dual immunofluorescence staining for CycT1 and pSer175 CDK9. Cell viability was also assessed by flow cytometry in three of the four experiments following staining with the eFluor450 viability dye. The viability data shown are expressed as a percentage relative to non-treated cells. (B) Combined inhibition of MEK and mTORC abrogates the expression of active P-TEFb. Memory CD4+ T cells (black) and resting primary Th17 cells (red) were treated or not for 30 min with the MEK inhibitor U0126 on its own or in combination with inhibitors targeting mTORC1 (Rapamycin) or both mTORC1 and mTORC2 (Torin) at the concentrations shown prior to TCR co-stimulation for 24 h. Thereafter, cells were examined for active P-TEFb expression by flow cytometry following dual immunofluorescence staining for CycT1 and pSer175 CDK9. In four of the experiments shown, cell viability was assessed by flow cytometry following staining with the eFluor450 viability dye. The viability data are expressed as a percentage relative to non-treated cells. Statistical significance (p values) in A and B was determined using a two-tailed Student’s t test. (C) Latently infected Th17 cells were treated or not for 30 min with the MEK inhibitor U0126 or the mTORC1/2 inhibitor Torin on their own or in combination prior to TCR co-stimulation for 24 h. Afterwards, cells were analyzed by flow cytometry following immunostaining using a fluorophore-conjugated antibody towards HIV Nef.
Fig 10.
Model for the reactivation of P-TEFb and proviral HIV by RasGRP1-Ras-Raf-MEK-ERK1/2 and PI3K-AKT-mTORC1/2 signaling in latently infected CD4+ T cells.
Generation of diacylglycerol (DAG) by activated PLC-γ1 immediately following TCR co-stimulation enables the recruitment of RasGRP1 to the plasma membrane, the initial activation of Ras via GTP loading and the allosteric activation of membrane-anchored SOS by GTP-bound Ras. This can also be accomplished by DAG-mimicking PKC agonists. Allosteric activation of SOS creates a positive Ras-GTP-SOS feedback loop that leads to maximal activation of Ras and formation of Ras-Raf complexes at the membrane. A phosphorylation cascade is triggered by Ras-Raf leading to activation of ERK1 and ERK2 which in turn stimulates the biogenesis of P-TEFb via mechanisms that are yet to be defined. RasGRP1 activity may be regulated by controlling the availability of DAG which, based on the RNA-seq data shown in S18 Fig, may get converted to phosphatidic acid (PA) primarily by DGK-α in primary T cells. TCR co-stimulation also rapidly stimulates PI3K leading to the activation of mTORC1 which may regulate P-TEFb biogenesis by enabling posttranscriptional synthesis of cyclin T1 (CycT1). Inhibitor-based experiments have clarified that intracellular calcium release and the activation of PKC are dispensable for formation of transcriptionally active P-TEFb. However, the recruitment of RNA polymerase II (RNAP II) to proviral HIV and its eventual phosphorylation by P-TEFb to stimulate processive transcription elongation are likely to be dependent on both Ca++ and PKC signaling.